[  ] Workshop Title

Bio-Driven X: Bio-inspired Design via Python and Machine Learning

​

[  ] Key Software / Tech to be taught

Python, Anaconda, GhPython, Add-on Developing

​

[  ] Workshop Description

According to any action and reaction from design to performance, DATA flows take on the flexibility and the way functions moves through geometry. Taking over such flows from different dimensions could possibly be done by carrying on Machine Learning algorithms. Since we use parametric modeling to get the most out of our design process; therefore, it assumes that the outcome would be a lot worth reliable by analysis to check how the related physical and mechanical parameters are getting involved in each iteration especially when it comes to design by features. Support Vector Machine, Gaussian Mixture and K-Means are among those algorithms we can use as techniques to have our evaluation process and Genotype/Phenotype Control over unyielding randomization procedures such as Genetic Algorithm. Correspondingly, what exactly could organize proportions between comprised parameters in any design to analysis model, defines different version yet adapted iteration to each scenario. This is the primacy behind the intelligent optimization to be empowered by our computational tools and high qualified algorithms. Every decision making process can now be correlated to the demanded performance generally and elaborately via hacking the DATA-Flows which addresses the design both in concept and detail. Thus, since ML can help us manipulating such DATA, demonstrate a process to select the optimization feature for surrogate our outputs based on, making an empiricism ambiance to learn how to deal with it, has been the main scope of this document behind the scene. Subsequent paragraphs, discuss the curriculum in a more elucidate delineation for each title.

​

Learning Python should be occurred within acquainted processes for designers

This title is up to be a retractable and more of an excel experience to design nature-wise by manipulating growth patterns data on free form structures. The whole process is to be commenced by Basic Python heading to how to use RhinoScriptSyntax, RhinoCommon and it follows using Anaconda headed to data implementation with Python Packages like Numpy and OpenCV. It defines an algorithm to read a nature pattern image’s pixels by RGB codes and tracks them with Alpha mode towards the setting of a surface subdivision. The later steps will be dedicated to having these data as a mesh points controllable by number sliders for the user to change the meshing parameters and have the process toward a 3d structure on free form surfaces as structural basis. The whole idea is meant to adapt free-form structure design based on patterns of nature, which we assume have the optimization and functional paradigms in each step of generations. It is obvious that appraising the performance of the iterations is necessary which is going to be handled either by Machine Learning or Genetic Algorithm.

​

The workshop is suitable for any designer wanting to gain a mentored experience in the emerging realm of Machine Learning in design.

​

[  ] Workshop Inclusions

• 2-Day Live ONLINE Workshop

• Live interaction & trouble-shooting with Tutor

• Example / Workshop Files for future access & practice

• Telegram Follow-up with Tutor for troubleshooting & collaboration opportunities

• 6 month access to recordings of all sessions from the online workshop for future reference & practice

​

[  ] Keywords 

#StructuralAnalysis #Optimization #Strength #GrowthPatterns #Nature #FreeForms #Machine Learning

​

[  ] Workshop Duration & Type

2 DAYS // 8 Hours on both days (incl. of breaks)

ONLINE // Interactive Zoom Workshop

 

[  ] Detailed Schedule

​

11:30 AM - 7:30 PM GMT

​

Day 1:

Session 01: Introduction to Workshop, Tutor's Presentations, Induction to Bio-Design

Session 02: Introduction to Python Coding, GhPython + RhinoCommon

Day 2:

Session 03: Hands on exercises with Python & Conda, Jupyter Notebook Env and Virtual Machine

Session 04: Collaborative Project with Python via OpenCV, Add-on tool-sets, Optimization Process

​

​